<source>
The default is false. With true the engine closes the connection where it would otherwise respond with a temporary refusal.
The manual names the purpose directly: this helps when the submitting program cannot handle 4xx responses. Such a program usually understands a dropped connection and tries again, while a temporary refusal it may take as final or get stuck in a loop on.
Temporary refusals at acceptance do not come out of nowhere: the node responds with them when it is busy — for example, when it has hit the incoming connection ceiling or the memory limit.
Investigation becomes more expensive on both sides because of this: your log shows a drop, theirs shows a drop, and they can be linked only by time.
What to set. The concession is needed rarely and selectively:
| Value | When |
|---|---|
| false (default) | the submitting side understands 4xx — the usual case |
| true | an old submission program that breaks or loops on 4xx |
The second is worth granting only to the source it was intended for: on a broad block it deprives even those who understood a clear response of it.
# a concession to one source, not to the whole node <source 10.0.0.5> disconnect-on-transient-error true</source>
what people get wrong and how it ends2 mistakes▶
our experience
§ 3.3.7.35
1 relation
grep -n 'disconnect-on-transient-error\|total-max-smtp-in\|smtp-await-slot' /etc/pmta/config
3 articles
◇How a message gets into the engineSix described submission methods: extended SMTP, four language bindings and a directory the node watches on disk.◇Who is allowed to submit mail and for whom to accept itOut of the box, the node accepts submission only from the machine itself; two different questions — “who can send through us” and “for whose addresses we accept mail” — are answered by different directives.◇Headers the sender uses to control deliverySix control headers let the submitting side choose a stream, tag a message with a job, set signing options and the key, the envelope identifier and the schedule; each is allowed separately and per source.